Multiphase flow meter case study using artificial neural network in petroleum industry

Wahyu Sasongko Putro, Maryam Molsadat Alaei, Andra Supriyanto, Agus Sulistyo Budi


An Artificial Neural Network (ANN) method can provide estimation product especially to design and analyze multiphase flow smart control in the petroleum industry. The conventional measurement of multiphase flow is very difficult due to mixing flow rate total that are complicated in output inlet and outlet pressure Separator (FSB – V – 04) Foxtrot well Platform, Pertamina Hulu Energi (PHE) West Java Indonesia. Based on the problem, this study aimed to estimate Flow Rate Gas (Qg), Flow Rate Oil (Qo) and Flow Rate Water (Qw) over 2012 and 2013. The observation data was taken from Department of Engineering Construction Pertamina Hulu Energi (PHE) Offshore North West Java (ONWJ) Indonesia in eight-month observation. The Multilayer Perceptron (MLP) was used in ANN architecture to obtain training result. Multi-input and multi-output (MIMO) structure and Levenberg Marquardt (LM) algorithms are suggested to process the data. The result showed the Root Mean Square Error (RMSE) value of output estimation reached 6.33E10 and 98.8 % of Variance Accounted For (VAF) during training section


Artificial Neural Network (ANN); Pertamina Hulu Energi (PHE) Offshore North West Java (ONWJ) Indonesia; Multiphase flow meter; Separator

Full Text:



M.A. Adewumi, J. Zhou, PSIG Annual Meeting, New Mexico, 1995.

J. Zhou, M.A. Adewumi, J. Societty of Petrolium Engineers 11 (1996)

H. Seraj, R. Mohd Fuaad, K. Marzuki, Y. Rubiyah, T. Iliya Tizhe,

International Journal on Smart Sensing and Intelligent Systems, 3

(2000) 4.

S. Wang, PhD Thesis, Univ of Tusla, 2000.

S. Bolat, O. Kalenderli. IJCI Proceedings of Intl. XIII Turkish

Symposium on Artificial Inteilligence and Neural Networks, 1 (2003)

H. Yu, B.M. Wilamowski. Levenberg-marquardt training, 2011.

M.H. Trauth, G. Robin, M. Norbert, S. Elisabeth. MATLAB recipes

for earth sciences, 2007.

W. Suparta, S.P. Wahyu, Journal of Physics: Conference Series 539

(2014) 012021.


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

e-ISSN No: 2356-2536
p-ISSN No: 2528-6854

Supported by :

Jurnal Online
Management Jurnal Elektronik

Pusat Dokumentasi dan Informasi Ilmiah
Lembaga Ilmu Pengetahuan Indonesia

Indexed by :